Not really thrilled with the current plot.
You’re setup is essentially 3 separate drivers license photos in action.
You might think of COMBINATIONS of people – which lets them interact more natrually.
First, which camera has the establishing wide shot. Knowing nothing about background and lighting and the hundred other things that can effect this – with nothing but the diagram to go on – I’d say your wide shot would balance best from between the jib and the Camera B position. Then pull out Cam 1 so that they have an angle on more than just “head on C” – you can get both an alternate 3 shot – a 2 shot of B+C as well as a single shot on C.
That leaves 3 to cover A&B for singles and the combo 2-shot.
All of this might change, however, if you have to SHOW THE MONITOR as well? If so, you MUST leave a shot so that we can see how the interviewees react to that – which might take a camera by itself – particularly if you want to orient the watchers in positions where you may have to “cross the line” to see the monitor/footage.
Just some thoughts.